aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Combine the direct and send mixersChris Robinson2014-06-139-260/+100
|
* Combine some dry and wet path typesChris Robinson2014-06-138-99/+84
|
* Pre-apply the crossfeed filter gain to the input sample coefficientsChris Robinson2014-06-122-26/+16
|
* Use floats for the BS2B filterChris Robinson2014-06-112-44/+44
|
* Make bs2b_cross_feed inlineChris Robinson2014-06-102-45/+35
|
* Don't try to set fp precision with __control87_2 or _controlfpChris Robinson2014-06-081-10/+4
|
* Print CPU extensions that are unavailable but requestedChris Robinson2014-06-081-6/+6
|
* Open and close the file in the main thread in alffplayChris Robinson2014-06-081-26/+28
| | | | | This avoid problems with the file being closed while a video refresh is still scheduled.
* Require SSE for SSE2, and SSE2 for SSE4.1Chris Robinson2014-06-071-2/+2
|
* Update the disable-cpu-exts config option descriptionChris Robinson2014-06-071-4/+5
|
* Update a couple sources with the proper authorChris Robinson2014-06-062-2/+2
|
* Move InitiatePositionArrays to mixer_defs.hChris Robinson2014-06-064-45/+21
|
* Add SSE2 and SSE4.1 linear resamplersTimothy Arceri2014-06-069-16/+294
| | | | | Currently the only way SSE 4.1 is detected is by using __get_cpuid, i.e. with GCC. Windows' IsProcessorFeaturePresent does not report SSE4.1 capabilities.
* Avoid a loop when updating the source position variablesChris Robinson2014-06-023-10/+14
|
* Add an ffmpeg-based example playerChris Robinson2014-06-023-1/+1692
| | | | | | This example shows proper timing and A/V synchronization with an OpenAL stream. Requires the in-progress AL_SOFTX_source_length extension to avoid manually keeping track of the buffer queue length.
* Only define struct timespec if _TIMESPEC_DEFINED isn't setChris Robinson2014-05-311-1/+7
| | | | This matches what the mingw-w64 headers define with the struct.
* Add extern "C" to common headers.Chris Robinson2014-05-294-0/+30
|
* Add methods to exchange and compare-exchange RefCount valuesChris Robinson2014-05-272-0/+40
|
* Avoid unnecessary local variablesChris Robinson2014-05-271-6/+6
|
* Implement condition variables for WindowsChris Robinson2014-05-272-10/+170
|
* Add an extension string for the in-progress source length extensionChris Robinson2014-05-261-1/+2
| | | | | This is the extension that allows you to get the full length of the source's buffer queue.
* Implement condition variables (POSIX only!)Chris Robinson2014-05-262-4/+55
| | | | | Windows requires Vista or newer to get the CONDITION_VARIABNLE API, but we currently only require XP.
* Don't print an ERR if pulse fails to get latency info due to no dataChris Robinson2014-05-261-3/+8
| | | | It just means it was called too quickly after starting.
* Add an option to get the length of a source's full queueChris Robinson2014-05-251-0/+95
| | | | | This simplifies keeping track how much a source has buffered in its queue, which reduces a bunch of unnecessary book keeping the app would have to do.
* Don't resume the backend if there's no contextsChris Robinson2014-05-241-10/+11
|
* Move the active source's offset out of the direct paramsChris Robinson2014-05-243-4/+4
|
* Update the examples to use SDL2Chris Robinson2014-05-234-301/+499
| | | | | | This should make it easier to build the examples on 64-bit, where there are no premade dev libs for SDL1 on 64-bit mingw. FindSDL_sound.cmake has been updated to accept SDL or SDL2.
* Mark a few more functions as constChris Robinson2014-05-233-12/+12
|
* Rename CONST_FUNC and PRINTF_STYLE, and fix non-GNU AL_PRINTChris Robinson2014-05-231-10/+10
|
* AL_SOFT_block_alignment is now considered doneChris Robinson2014-05-223-7/+7
|
* Ensure a proper amount of new elements are being reserved when insertingChris Robinson2014-05-221-1/+5
|
* Include strings.h if it exists for strncasecmpChris Robinson2014-05-221-1/+3
|
* Add a needed header for the CoreAudio backendChris Robinson2014-05-221-0/+1
|
* Use the first non-0-length buffer when starting a sourceChris Robinson2014-05-221-1/+1
|
* Don't try to use the non-standard alloca.hChris Robinson2014-05-221-8/+1
| | | | | Variable-length arrays are C99 standard and are preferred over alloca. The only compiler that needs alloca (MSVC) has it in malloc.h.
* Use an unsigned type for the win32 size_t formatterChris Robinson2014-05-221-1/+1
|
* Mark some functions as constChris Robinson2014-05-221-5/+7
|
* The lower value of the gain vector contains the closest target valueChris Robinson2014-05-211-2/+2
|
* Don't clear the current and step gain values when updating a sourceChris Robinson2014-05-211-89/+66
|
* Use the current buffer item for static sources tooChris Robinson2014-05-211-1/+1
|
* Copy samples if needed in the 'copy' resamplerChris Robinson2014-05-211-9/+15
|
* Properly process all samples with a bandpass filterChris Robinson2014-05-211-2/+2
|
* Use the correct input for HRTF mixingChris Robinson2014-05-211-1/+1
|
* Update the active source's direct and send properties after each mixChris Robinson2014-05-211-10/+9
|
* Add some asserts to verify block alignment for conversionChris Robinson2014-05-201-0/+8
|
* Return a sample pointer from resamplersChris Robinson2014-05-195-24/+27
| | | | Both resampling and filtering now avoid copying samples when they no-op.
* Put per-channel filter properties togetherChris Robinson2014-05-193-39/+46
|
* Return a sample pointer from DoFiltersChris Robinson2014-05-191-14/+18
| | | | | This allows us to avoid copying samples when no filtering is needed. Doing the same for resampling would require another temporary buffer in the device.
* Move an HRTF mixer parameter and shorten a couple variable namesChris Robinson2014-05-184-33/+28
|
* Don't pass the SendParams to the wet-path mixerChris Robinson2014-05-186-41/+44
|